As in the case of Sterecon, the first step in beginning a tomographic study is to clarify precisely what the investigator hopes to learn from a reconstruction. We may find that a simple tilt-series animation, or the use of Sterecon, would be more appropriate. Once we agree that a tomographic study is appropriate, the choice of specimen thickness and magnification have to be carefully considered. These decisions are discussed by the user and all three of the listed investigators, with Dr. McEwen making the final recommendation to the user. The user is then trained in recording the tilt-series in the HVEM (usually by Buttle) or the IVEM (usually by Marko). In some cases, our staff does the microscopy. Protection of the specimen from contamination and excessive irradiation are important here. The recording of a single-axis tilt series, typically takes 2-3 hours, but we normally do double-tilt recording, which takes twice as long. If images are recorded on film, we assi st the user in setting up the scanning camera for scanning and pre-alignment. The user then completes the scanning in about 2 hours for a double-tilt series (122 images). The final alignment uses gold markers as fudicial marks, and must be done even if the images are recorded digitally using the IVEM. The locations of the gold markers must be indexed on each image. This procedure takes about 3 hours, and only requires a few minutes of training. The reconstruction can be completed in about 3 hours for a double-tilt volume of about 512x512x200 pixels, or half an hour for a single-tilt volume of the same size. Visualization of the reconstruction is the most time-consuming portion of the project. In many cases, we do this with the investigator looking on. If the investigator is to do this independently, there must be training in the use of our workstations, the SPIDER software system, and VoxelView, or another visualization package. In most cases Sterecon is used for segmentation , the first step in visualization. This training takes about a week. Some investigators have visualization capabilities in their own labs, and if they use software we are familiar with (or supply), we also provide assistance, usually over Internet. We take every opportunity to promote the unique benefits of HVEM tomography at meetings via presentations and informal discussions.